Individual differences in colourfulness judgments of images of natural scenes

The purpose of this study is to estimate individual differences in colourfulness judgments of four real-life scenes by means of direct scaling and difference scaling. Images were created by varying chroma in the CIELUV colour space while lightness and hue were kept constant. The results indicate that the strategy of colourfulness judgments varies among observers: some subjects use one unified scale and score the colourfulness on the basis of the absolute values of average chroma and its standard deviation, while others use several scales and score the colourfulness on the basis of the relative values. i.e., differences from a scene-dependent reference value, of average chroma and its standard deviation separately per scene. The difference-scaling procedure corresponds more with using one unified scale, and the direct-scaling procedure with using separate scales. A model describing this subjective bias is presented.